This study will evaluate the effect of a well-defined physical intervention (constraint-induced movement therapy (CIMT)) on cortical motor reorganization following stroke using transcranial magnetic stimulation (TMS) and functional magnetic resonance imaging (fMRI). The primary objectives are 1) to determine whether TMS motor maps or fMRI activation predictably change following CIMT, 2) to assess whether this change correlates with motor recovery, and 3) to identify possible mechanisms underlying that change. Eighty patients participating in a randomized multicenter trial (EXCITE) of CIMT administered either immediately or one year later will be studied. Clinical assessment, TMS and fMRI will be obtained prior to randomization into the treatment arms, and repeated within three days of completing CIMT (or at the equivalent time for patients randomized to delayed treatment), and one year later. The primary focus of the study will be changes in the primary motor cortices (FMC's) of each hemisphere in response to CIMT, but we will also qualitatively evaluate activation changes in other motor regions. Outcome measures for TMS will include indices of cortical map area and center of gravity to evaluate changes in size and directional shift of motor maps over time. Functional MRI outcome measures will be generated from a volume of interest (VOI) analysis of activation in the PMC during a motor task, and will parallel TMS measures. In addition, a "laterality index" will measure the relative activation in the PMC's contralateral and ipsilateral to hand movement. Clinical outcome measures will be obtained as part of the EXCITE trial, and will include the laboratory-based Wolf Motor Function Test and a measure of real life use of the stroke-affected arm: the Motor Activity Log. Nonparametric statistical analyses will be used to examine the relationship between changes in TMS maps, fMRI activation and clinical outcome for each group of patients. The study is a unique opportunity to use non-invasive methods to study cortical plasticity in subjects receiving a specific therapy intervention, compared to a control group. We expect to identify specific patterns of change in TMS motor maps and fMRI activation patterns in response to the CIMT intervention. The project will validate the concept that physiological cortical motor changes after stroke are closely correlated with motor improvement and are influenced by a physical intervention. This could lead to the development of new rehabilitative strategies based on the interaction between therapeutic interventions and the physiology and anatomy of recovery.